Nearly 195,800 men newly diagnosed with localized prostate cancer (PCa) in the United States each year face difficulty in deciding whether potentially harmful PCa treatment is necessary due to our inability to accurately distinguish between indolent and aggressive PCa. Established prediction nomograms based on routine clinical and pathological features of PCa misclassify up to 60% of men with clinically significant cancer, most likely because these models do not effectively account for the myriad of biological mechanisms that can underlie metastatic potential. For the same reason, these existing prediction tools offer little insight on potential molecular targets for PCa treatment. Recent studies have demonstrated that aberrant DNA methylation of numerous genes, including metastasis suppressors and genes that maintain cell differentiation status, is a major mechanism underlying tumor metastatic progression. We hypothesize that tumor DNA methylation of specific genes are predictive of PCa metastatic progression beyond routine clinical and pathological prognostic factors. The overarching goal of this proposal is to develop DNA methylation marker-based prediction algorithms that will assist clinicians when providing therapeutic recommendations for men with localized PCa. Existing PCa study populations consist predominately of treated patients. They are not well-suited to address whether tumor DNA methylome measureable at PCa diagnosis is an important risk factor for metastasis because PCa treatment can alter disease course in at least some of these men. To evaluate the prognostic utility of DNA methylation markers measurable from diagnostic PCa biopsy, we propose a nested case-control study based on a retrospective cohort of 9,563 men who did not receive treatment when diagnosed with localized, Gleason grade 6-7 PCa (1997-2006) at Kaiser Permanente (KP). We will include 450 cases with metastasis and 450 matched controls selected during a 10-year follow-up period. The Specific Aims are: (1) to test the hypothesis that DNA methylation status of selected candidate genes can predict risk of PCa metastasis; (2) to identify novel DNA methylation markers of genes with previously unrecognized utility in predicting PCa metastasis from genome-wide methylation profiling using a discovery approach; and (3) to develop prediction algorithms of the 10-year risk of metastasis in men with localized PCa based on existing clinical PCa prognostic criteria and subsequently incorporating DNA methylation markers. This application takes advantage of unique resources in place at KP, University of California Los Angeles and San Francisco, and University of Southern California, leveraging a multidisciplinary team of investigators with expertise in prognosis studies, PCa epidemiology, epigenetics, pathology, clinical urology, and biostatistics. At the completion of these aims, we will identify DNA methylation markers that are predictive of PCa metastasis, provide insights on the molecular determinants of metastatic progression, and develop novel risk algorithms that extend existing algorithms to metastasis. 1